Solinftec

Award for Innovation

2024
Our main use case is real-time GPS correction for our autonomous robots, using Real-Time Kinematic. With data streaming powered by Apache Kafka, we guarantee millimeter precision in the field.
Data streaming powered by Apache Kafka guarantees millimeter precision in the field.

Our main use case is real-time GPS correction for our autonomous robots, using Real-Time Kinematic. With data streaming powered by Apache Kafka, we guarantee millimeter precision in the field.

About the company:

Founded in Brazil, Solinftec is a global leader in artificial intelligence and robotics for agribusiness, with over 70,000 machines in the field and more than 30 million acres of land optimized using our solutions.

By leveraging IoT, AI, and data streams, Solinftec enhances on-farm operations management and agronomic decision-making through detailed field-level data. The company digitizes farming operations by integrating software and hardware solutions, enabling real-time diagnosis and action.

Recently, Solinftec introduced Solix, an autonomous robotic platform that identifies weeds and applies herbicides in real time. This innovation reduces herbicide usage by approximately 90% and increases productivity through more frequent and precise management, resulting in an additional 10 bags per hectare in the last season in Brazil and United States. Beyond improving efficiency, Solix plays a crucial role in advancing the sustainability of global food production by minimizing chemical inputs and promoting more responsible and environmentally friendly farming practices.

Data Streaming Technology Used:

Apache Kafka, Apache Flink, Apache Pinot and Superset, Spring Boot, UDP protocol

What problem were they looking to solve with Data Streaming Technology?

For some cultures, the plants are sown in rows in the fields and those rows frequently are not very far away from each other. All machines that work in the field must be cautious to not crush the growing plants. Since our robots are autonomous we need to ensure that our GPS coordinates are precise to the millimeter scale in order for the Solix bots to both not trample the plants and have exact precision while geolocalization the weeds and spraying herbicides.

How did they solve the problem?

A base station at a fixed location receives signals from GPS satellites and calculates errors caused by factors like the ionosphere, troposphere, and satellite orbits. It sends these differential corrections to a mobile rover via a real-time data link. Using Apache Kafka, the corrections are continuously transmitted with low latency. The rover applies these corrections to its position data, resulting in centimeter or millimeter level accuracy.

What was the positive outcome? 

Apache Kafka provides a scalable solution that can handle high throughput of data. Overall, the integration of data streaming technology in RTK GPS systems has resulted in significantly improved positioning accuracy, efficiency, and reliability.

Additional links:

https://www.solinftec.com/en-us/

https://www.solinftec.com/en-us/alice-ai-solix-ag-robotics-2/